Update README.md

README.md

@@ -75,6 +75,10 @@ ASSISTANT:
 - Capybara is trained mostly on multi-turn, long context, highly curated and cleaned GPT-4 conversations with real humans.
 
 For these reasons, it's reccomended to give Capybara a try if you want to have multi-turn conversations and/or long context communication.
+
+## Benchmarks! (Important to note that all the mentioned benchmarks are in a single-turn dataset setting, Capybara should excel even further at multi-turn conversational tasks.)
+
+![Capybara](https://i.imgur.com/n8lkmyK.png)
  
 ## Limitations
 
@@ -112,6 +116,3 @@ The following are benchmarks we checked for contamination for:
 
 - GPT4All
 
-## Benchmarks! (Important to note that all the mentioned benchmarks are in a single-turn dataset setting, Capybara should excel even further at multi-turn conversational tasks.)
-
-![Capybara](https://i.imgur.com/n8lkmyK.png)